PRESS-NEWS.org - Press Release Distribution
PRESS RELEASES DISTRIBUTION

Integrating machine learning with statistical methods enhances disease risk prediction models

Integrating machine learning with statistical methods enhances disease risk prediction models
2024-10-14
(Press-News.org)

Researchers from Peking University have conducted a comprehensive systematic review on the integration of machine learning into statistical methods for disease risk prediction models, shedding light on the potential of such integrated models in clinical diagnosis and screening practices. The study, led by Professor Feng Sun from the Department of Epidemiology and Biostatistics, School of Public Health, Peking University, has been published in Health Data Science.

Disease risk prediction is crucial for early diagnosis and effective clinical decision-making. However, traditional statistical models, such as logistic regression and Cox proportional hazards regression, often face limitations due to underlying assumptions that may not always hold in practice. Meanwhile, machine learning methods, despite their flexibility and ability to handle complex and unstructured data, have not consistently demonstrated superior performance over traditional models in certain scenarios. To address these challenges, integrating machine learning with traditional statistical methods may offer more robust and accurate prediction models.

The systematic review analyzed various integration strategies for classification and regression models, including majority voting, weighted voting, stacking, and model selection, based on whether predictions from statistical methods and machine learning disagreed. The study found that integration models generally outperformed both statistical and machine learning methods when used alone. For example, stacking was particularly effective for models involving over 100 predictors, as it allows for the combination of the strengths of different models while minimizing weaknesses.

“Our findings suggest that integrating machine learning into traditional statistical methods can provide more accurate and generalizable models for disease risk prediction,” said Professor Feng Sun, the lead researcher. “This approach has the potential to enhance clinical decision-making and improve patient outcomes.”

Looking ahead, the research team plans to validate and improve existing integration methods further and develop comprehensive tools for evaluating these models in various clinical settings. The ultimate goal is to establish more efficient and generalizable integration models tailored to different scenarios, ultimately advancing clinical diagnosis and screening practices.

END


[Attachments] See images for this press release:
Integrating machine learning with statistical methods enhances disease risk prediction models

ELSE PRESS RELEASES FROM THIS DATE:

Changing watering practices to improve tomato plant health

Changing watering practices to improve tomato plant health
2024-10-14
Some people believe that talking to your plants makes them thrive. While there’s limited scientific support for sound improving plant health, there’s a growing amount of evidence about the benefits of mechanical stimulation, like touch, wind or rain. Researchers reporting in ACS’ Journal of Agricultural and Food Chemistry examined the impact of watering practices on tomato plants. They found that the size of the water droplets affected plant growth and resistance to pests and pathogens. Climate change threatens crop production as rising ...

Six proteins implicated in early-onset preeclampsia

2024-10-14
Preeclampsia is a life-threatening pregnancy complication marked by persistent high blood pressure that is even more serious when it occurs early in the first trimester. The exact cause of early-onset preeclampsia is unknown, and it is difficult to predict, prevent and diagnose. Now, in ACS’ Journal of Proteome Research, researchers report on six proteins that could be used as targets to diagnose and treat the condition. Preeclampsia’s key symptom is high maternal blood pressure, and serious cases can lead to maternal organ failure, low infant birth weight, or maternal or fetal death. ...

Researchers at The University of Texas at San Antonio reveal oversight in AI image recognition tools

Researchers at The University of Texas at San Antonio reveal oversight in AI image recognition tools
2024-10-14
Artificial intelligence can help people process and comprehend large amounts of data with precision, but the modern image recognition platforms and computer vision models that are built into AI frequently overlook an important back-end feature called the alpha channel, which controls the transparency of images, according to a new study. Researchers at The University of Texas at San Antonio (UTSA) developed a proprietary attack called AlphaDog to study how hackers can exploit this oversight. Their findings are described in a paper written by Guenevere Chen, an assistant professor in the UTSA Department of Electrical and Computer Engineering, and her former ...

World of crayfish™: A web platform for global mapping of freshwater crayfish and pathogens

World of crayfish™: A web platform for global mapping of freshwater crayfish and pathogens
2024-10-14
Freshwater crayfish, keystone species of aquatic ecosystems, are now at the center of a new tool that promises to revolutionize conservation efforts. World of Crayfish™ (WoC), a newly launched web platform, aims to provide real-time global mapping of freshwater crayfish and their pathogens, offering an invaluable resource for researchers, conservationists, and environmental stakeholders. Published in PeerJ Life and Environment, the article, World of Crayfish™: A web platform towards real-time global mapping of freshwater crayfish and their pathogens, highlights the urgent need for accurate, up-to-date data on ...

How to make biodiversity credits work: science-based solutions for real conservation gains

2024-10-14
Biodiversity is in crisis. Human activities are driving species extinctions at unprecedented rates, but funding for conservation remains woefully inadequate. To address this gap, the concept of a Biodiversity Credit Market (BCM) has emerged, inspired by carbon credit systems that incentivize conservation and restoration efforts.  However, while the BCM holds promise, it risks falling into the same pitfalls as its carbon counterpart—especially if it fails to implement rigorous science-backed baselines, ...

Qunova becomes first to achieve ‘chemical accuracy’ on commercial quantum computers with its hardware agnostic algorithm

Qunova becomes first to achieve ‘chemical accuracy’ on commercial quantum computers with its hardware agnostic algorithm
2024-10-14
DAEJEON, South Korea (October 14, 2024) – Qunova Computing, a developer of quantum software applications designed to bring quantum computing to the chemical, pharmaceutical and industrial engineering industries, today announces the results from a series of recent tests performed on three different NISQ era quantum computers, each with a different qubit count. In each demonstration, Qunova’s algorithm was able to produce results with accuracy below the threshold of 1.6 millihartrees required for real-world quantum chemistry applications, a level known as ‘chemical ...

Scientists have successfully bred corals to improve their heat tolerance

2024-10-14
Scientists have successfully bred corals to improve their heat tolerance A new study has shown that selective breeding can lead to a modest rise in coral heat tolerance. Led by experts at Newcastle University’s Coralassist Lab, the study documents the world’s first effort to selectively breed adult corals for enhanced heat tolerance, i.e. the ability of adult corals to survive intense marine heatwaves. The breeding effort was a success, showing that it is possible to improve the heat tolerance of adult coral offspring, even in a single generation. However, the improvement was modest in comparison ...

Adaptability of trees persists after millions of years of climate change

Adaptability of trees persists after millions of years of climate change
2024-10-14
Seven of the most common forest trees in Europe have been shown to be able to shelter their genetic diversity from major shifts in environmental conditions. This is despite their ranges having shrunk and the number of trees having fallen sharply during ice age cycles. These are the findings of a study by a European consortium including Uppsala University, published in Nature Communications. “From a biodiversity perspective, this is very positive because these trees are keystone species on which many other species depend,” says Pascal Milesi, Associate Professor of Plant Ecology and Evolution at Uppsala University and first author of the study. The researchers aimed to investigate ...

Protein involved in balancing DNA replication and restarting found

2024-10-14
A protein that is involved in determining which enzymes cut or unwind DNA during the replication process has been identified in a new study.   In a new paper published in Nature Communications, an international team of researchers have found that the protein USP50 supports the DNA replication process by helping to decide the proper use of nucleases or helicases. These enzymes are implemented during the DNA replication process to promote ongoing replication and where the copying machinery runs into problems ...

How liberals and conservatives can have better conversations, according to a psychologist

2024-10-14
Bridging Our Political Divide collects the insights of a psychologist who offers antidotes to the unproductive arguments that now dominate our political culture and ways to find common ground. New York, U.S./Oxford, UK, October 14, 2024 – As the 2024 US Presidential Election in November inches closer, the partisan divide in American society has never felt more apparent – or pervasive. From workplace chatter to heated discussions at family gatherings, many Americans may have asked themselves, ‘How can I make this ...

LAST 30 PRESS RELEASES:

Cheese may really be giving you nightmares, scientists find

Study reveals most common medical emergencies in schools

Breathable yet protective: Next-gen medical textiles with micro/nano networks

Frequency-engineered MXene supercapacitors enable efficient pulse charging in TENG–SC hybrid systems

Developed an AI-based classification system for facial pigmented lesions

Achieving 20% efficiency in halogen-free organic solar cells via isomeric additive-mediated sequential processing

New book Terraglossia reclaims language, Country and culture

The most effective diabetes drugs don't reach enough patients yet

Breast cancer risk in younger women may be influenced by hormone therapy

Strategies for staying smoke-free after rehab

Commentary questions the potential benefit of levothyroxine treatment of mild hypothyroidism during pregnancy

Study projects over 14 million preventable deaths by 2030 if USAID defunding continues

New study reveals 33% gap in transplant access for UK’s poorest children

Dysregulated epigenetic memory in early embryos offers new clues to the inheritance of polycystic ovary syndrome (PCOS)

IVF and IUI pregnancy rates remain stable across Europe, despite an increasing uptake of single embryo transfer

It takes a village: Chimpanzee babies do better when their moms have social connections

From lab to market: how renewable polymers could transform medicine

Striking increase in obesity observed among youth between 2011 and 2023

No evidence that medications trigger microscopic colitis in older adults

NYUAD researchers find link between brain growth and mental health disorders

Aging-related inflammation is not universal across human populations, new study finds

University of Oregon to create national children’s mental health center with $11 million federal grant

Rare achievement: UTA undergrad publishes research

Fact or fiction? The ADHD info dilemma

Genetic ancestry linked to risk of severe dengue

Genomes reveal the Norwegian lemming as one of the youngest mammal species

Early birds get the burn: Monash study finds early bedtimes associated with more physical activity

Groundbreaking analysis provides day-by-day insight into prehistoric plankton’s capacity for change

Southern Ocean saltier, hotter and losing ice fast as decades-long trend unexpectedly reverses

Human fishing reshaped Caribbean reef food webs, 7000-year old exposed fossilized reefs reveal

[Press-News.org] Integrating machine learning with statistical methods enhances disease risk prediction models